The long term goal of these studies is to develop a heart regeneration strategy that can be translated to humans. Our current focus is in human pluripotent stem cells, and more specifically, in human embryonic stem cell-derived cardiomyocytes (hESC-CMs). Building on 15 years' work in cell culture and small animals, we recently showed that hESC-CMs can be produced at a scale and purity that permit testing in the animal most likely to predict the human response: the non-human primate Macaca nemestrina. The hESC-CMs can generate grafts that average 40% of infarct size, and they electromechanically couple, beat in sync with the macaque heart and mature to adult size by 3 months. This proposal addresses the next-level challenges. Aim 1 is a pivotal efficacy study that will determine if hESC-CMs enhance contractile function in the infarcted macaque heart. Three doses of human cardiomyocytes will be tested, and ventricular structure and function will be determined by MRI. Aim 2 addresses the genesis of ventricular arrhythmias identified in our recent study. In addition to monitoring heart rhythm by telemetry, we will perform serial catheter-based mapping of the engrafted (3 cell doses) and sham-engrafted primate heart, generating activation maps that will classify arrhythmias as resulting from reentry or ectopic pacemakers. Programmed electrical stimulation will assess susceptibility to induced arrhythmias. In combination with Aim 1, these studies will determine if there is a dose of cardiomyocytes that enhances function without arrhythmias. Aim 3 uses molecular imaging techniques to assess graft population dynamics, tracking the waves of cell death and proliferation after transplantation. We also address the key question of graft vascularization, testing the hypothesis that grafts induce host coronary arterial remodeling and that, despite this response, that grafts are chronically underperfused. These experiments will provide the foundation for a planned clinical trial of heart repair and provide mechanistic insights not possible in human patients.